EP4621865A1 - Activating method for an all-solid-state battery comprising halide electrolyte - Google Patents

Abstract

The present invention relates to a method for activating an all-solid-state battery comprising a halide electrolyte having the formula:

        Li_{[(x-y)+a+b+c+p]}Na_yZr_(1-a-b-c)Er_aY_bSC_cCl_{[(4+x)-m-n-o-p]}I_mBr_nF_oO_p

in which: 0.2 ≤ x ≤ 6; 0 ≤ y ≤ 6; 0 ≤ a ≤ 1; 0 ≤ b ≤ 1; 0 ≤ c ≤ 1; 0 ≤ m ≤ 1; 0 ≤ n ≤ 1; 0 ≤ o ≤ 1; 0 ≤ p ≤ 2; with 0 ≤ a+b+c ≤ 1 and (4+x) > (m+n+o+p), characterized in that the battery is activated by charging it at a voltage greater than 4.0 V vs Li⁺/Li.

Description

• The present invention relates to a method for activating an all-solid-state battery comprising halide electrolyte.
• All-Solid-State Batteries (ASSBs) stand poised to surpass the energy densities of Li-ion batteries and alleviate safety concerns. Among diverse solid-state electrolytes (SEs), halides, notably Li₂ZrCl₆ (LZC), exhibit stability window up to 4 V vs Li⁺/Li⁰ and whereas Y-doped LZC showed a reversible electrochemical activity in this potential region. Yet, the untapped potential of halides SEs for enhancing overall ASSB performance remains unexplored, diverging from their conventional use solely for ionic conduction.
• It has been discovered that halide electrolytes under certain activation conditions allow remarkable enhancements in the specific capacity of all solid-state cells driven by combined effects that concurrently address major issues in silicon and anode-less chemistry. Notably, such activation of the battery yields outstanding performance from industry-relevant, highly loaded electrodes, showcasing significant advancements in ASSB technology. Furthermore, the tunability of these effects by adjusting the initial Li composition of the solid electrolyte (SE) is demonstrated.
• Thus, the invention relates to a method for activating an all-solid-state battery comprising a halide electrolyte having the formula:
  
          Li_{[(x-y)+a+b+c+p]}Na_yZr_(1-a-b-c)Er_aY_bSc_cCl_{[(4+x)-m-n-o-p]}I_mBr_nF_oO_p
  
  in which:
• $$0.2\le \mathrm{x}\le 6;$$
• $$0\le \mathrm{y}\le 6;$$
• $$0\le \mathrm{a}\le 1;$$
• $$0\le \mathrm{b}\le 1;$$
• $$0\le \mathrm{c}\le 1;$$
• $$0\le \mathrm{m}\le 1;$$
• $$0\le \mathrm{n}\le 1;$$
• $$0\le \mathrm{o}\le 1;$$
• $$0\le \mathrm{p}\le 2;$$
• with 0 ≤ a+b+c ≤ 1,
• with (4+x) > (m+n+o+p),
characterized in that the battery is activated by charging it at a voltage greater than 4.0 V vs Li⁺/Li.
• In one particular embodiment, the battery can be activated by charging it at a voltage greater than or equal to 4.5 V vs Li⁺/Li, preferably greater than or equal to 5.0 V vs Li⁺/Li.
• The halide electrolyte may be selected from: Li_xZrCl_4+x, wherein x is a number from 0.4 to 6, being in particular Li_0.4ZrCl_4.4, Li_0.59ZrCl_4.59, Li_0.8ZrCl_4.8, LiZrCl₅, Li₂ZrCl₆, Li₃ZrCl₇, Li₄ZrCl₈, Li₅ZrCl₉ and Li₆ZrCl₁₀.
• In one particular embodiment, the battery has a cathode, the active material of which can be selected from LiFePO₄, LiMn₂O₄, Li₃V₂(PO₄)₃, LiVOPO₄, Li₂VOPO₄F and Li_a'Ni_x'Mn_y'Co_z'O₂ with x'+y'+z'=1 and 0 < a' < 1.2.
• In one particular embodiment, the battery is anode less.
• In one particular embodiment, the battery has an anode, the active material of which can be selected from lithium metal, lithium-indium alloy, silicon, tin or carbon intercalation materials. The lithium-indium alloy can be LiIn_x" with 0 < x" ≤ 3. The intercalation materials can be selected from graphite and hard carbon.
• In one particular embodiment, the halide electrolyte can be under the form of a layer. The battery can also further comprise a second electrolyte layer comprised of sulfide electrolyte, such as Li₆PS₅Cl.
METHODS: Electrolytes synthesis:
• Li_xZrCl_4+x (x = 0. 4, 0. 59, 0. 8, 1, 2, 3) (LZC) electrolytes were synthesized using starting materials: LiCl (>99%, Sigma Aldrich), dried overnight at 160°C under vacuum, and ZrCl₄ (98%, anhydrous, Sigma Aldrich).
• The stoichiometric amount of LiCl was first crushed in a mortar with pestle in dry argon glove box, to reduce the particle size. After that, the stoichiometric amount of ZrCl₄ was added to the mortar and mixed for 10 minutes with LiCl. The mixture was then placed in a 25ml ZrO₂ ball-milling jars along with 10 ZrO₂ balls of 10 mm diameter. The Li_xZrCl_4+x electrolytes used in this work were synthesized over 50h at 600 rpm, with 15 minutes break between each 1 hour cycle. XRD diagrams of Li_xZrCl_4+x compounds along with the full pattern matching refinement of Li₂ZrCl₆ that confirms the synthesis steps are reported in Figure 1. Asterisks in Figure 1b indicate the presence of a new set of broad peaks vs Li₂ZrCl₆. Variations of the Li_xZrCl_4+x conductivities vs x has been characterized by impedance spectroscopy (Figure 2).
Electrode preparation: Cathodes
• The cathodes used for the evaluation of the electrolyte redox activity, that is the ones consisting of Li_xZrCl_4+x electrolyte and single wall carbon nanotubes (P2SWNT), as well as the electrodes with LiFePO₄ (LFP) as the active material and vapor grown carbon fibers (VGCFs), used for galvanostatic cycling, are prepared by mixing the materials in their respective weight ratios by ball milling at 300 rpm for 2 hours. The ball milling was done in the same system described for the case of electrolytes synthesis.
Anodes
• Li_0.33In composite anode was prepared by folding Li metal foil into an In metal foil in 2:98 weight ratio, until the resulting Li_0.33In alloy became brittle. After that the foil was crushed and mixed using mortar and pestle with argyrodite Li₆PS₅Cl (LPSC) electrolyte (NEI Corporation) in 60:40 wt% ratio.
• To make Si composite anode, (LPSC) and Si powder having grain size in the micrometer range (referred to as micro-Si) were mixed in the ball milling jar in 50:50 under the same conditions as for the cathode.
Electrochemical impedance spectroscopy:
• To measure the conductivity of electrolytes, 120 mg of electrolyte powder was pressed under 375 MPa for 3 minutes, between two titanium plungers in an in-house built system comprised of 2 Ti plungers and a PEEK insulating cylinder. Then, the cell was taken into a pressure holder that exerted 100 MPa of stack pressure onto it.
• To measure the impedance of the full cell with different composite cathodes, the EIS measurements were taken after each cycle or after each charge or discharge sequence at the same pressure used during the cycling of the cells.
• The sinusoidal voltage of 10mV amplitude was applied in the frequency range 100 mHz - 100 kHz to the cell. The corresponding signal is plotted in the form of Nyquist diagram, typically showing a straight line. The resistance of the electrolyte was found by the intersect of the interpolated straight line with the real axis. The conductivity was calculated based on the equation: $$\sigma =\mathrm{l}/\mathrm{A}*1/\mathrm{R},$$ where l is the thickness of the electrolyte pellet (measured individually for each pellet: usually in the 500-700 µm range) and A is the surface area of the pellet (0.785 cm²)
Cells assembly:
• Each cell was made in the in-house made system comprised of 2 Ti plungers and a PEEK insulating cylinder. By compressing 40mg layer of Li₆PS₅Cl (NEI Corporation) at 125MPa for 10 seconds, the first layer (LPSC layer) of separator is formed. Second layer of the bi-layer separator is made by compressing 40mg Li_xZrCl_4+x (x = 0. 4, 0. 59, 0. 8, 1, 2 and 3) (LZC layer) on top of the first layer at 375 MPa for 1 minute. Then, appropriate amount of specific composite cathode powder was placed on top the LZC layer, and corresponding amount of composite anode, when present, was placed on top of the LPSC layer. Finally, the whole stack was pressed at 375MPa for 3 minutes. Each cell would, then, be placed in an in-house designed holder, and the pressure of either 100 MPa or 75 MPa was applied.
• The three cell schematics are depicted in Figure 3. Each of the cell integrates a positive electrode resulting from the mixture of LZC, LFP and VGCF, a LZC/LPSC bilayer electrolyte and Li_0.33In anode in cell (a), anode-less in cell (b) and Si anode in cell (c).
Cyclic voltammetry:
• Around 12 mg of composite cathode composed of Li_xZrCl_4+x (x = 0.4, 0.59, 0.8, 1, 2 and 3) and single wall carbon nanotubes referred to as P2SWNT (Carbon solution inc. ) in 85:15 wt% was used for each measurement. The voltage was linearly increased/decreased in the range 2.5V - 5V vs Li+/Li, while the current was measured. Each experiment was repeated 6 times. To evaluate the capacity the area under current (mA/g) - time (h) was integrated using Origin software or EC-Lab software.
Full cell cycling:
• Full cell with Li_0.33In alloy composite anode was made using 18 mg of composite cathode made out of LZC, LFP and vapor grown carbon fibers (VGCFs) in a 59.7:33.3:7 weight ratio as described previously, and 30 mg of the composite anode.
• Cell was cycled galvanostatically at C/10 for 2 cycles: during the first cycle the voltage range was 2.5V - 3.8 V vs Li+/Li in order to extract the capacity coming purely from LFP cathode active material; on the second cycle the voltage range of 2.5V - 5V vs Li+/Li was applied in order to utilize the capacity of LFP and oxidation of LZC electrolyte. This was followed by 5 cycles at C/5, and C/2, after which 100 cycles at 1C were performed in the voltage range 2.5 V - 5 V vs Li⁺/Li. Finally, the capacity retention of both LFP and LZC oxidation after 100 cycles is demonstrated by performing the same cycling conditions as in the initial 2 cycles.
• Anode-less cell was utilizing 11.5 mg of composite cathode made out of LZC, LFP and VGCF in 53.7: 33.3:10 weight ratio in order to extract more capacity from LZC oxidation and improve the performance of the anode-less cell. Galvanostatic cycling was performed at C/10 for 11 cycles.
• Full cell using a composite silicon anode was composed of 43.5 mg of composite cathode consisting of LZC, LFP and VGCF in 44.7:50:5.3 weight ratio and composite anode made of the mixture of LPSC and micro-Si as previously described, while having the nominal capacity of the anode set to be 3 times larger than that of the cathode. This was done to insure most of the capacity fade is a result of interphase (referred to as SEI) formation between micro-Si and LPSC electrolyte. The full cell was, then, galvanostatically cycled in 2.5 V - 3.8 V vs Li⁺/Li voltage ratio at C/10 (C being the theoretical capacity of the cathode). After that, the cell undergoes 1 cycle in the voltage range 2.5 V - 5V in order to extract the capacity from the LZC oxidation, followed by 3 cycles in lower voltage range, to unveil if this additional capacity could increase the Li⁺ ion storage in the LFP cathode material.
RESULTS
• The cells integrating positive electrode resulting from the mixture of LZC, LFP and VGCF, a LZC/Li₆PS₅Cl bilayer electrolyte and Li_0.33In, anode-less or micro-Si negative electrode as represented in Figure 3 have been tested. These cells were charged either to 4V, which is typical when cycling LFP active material, or 5V to purposely oxidize the halide electrolyte.
• For the Li_0.33In half-cell, under a 4 V charge, the typical LFP profile delivers nearly 150 mAh/g_LFP, (50 mAh/g_cathode) under 0.12 mA/cm² (Figure 4). However, when the charge is extended to 5V, the LZC electroactivity occurs from 4 V vs Li⁺/Li, dramatically enhancing the overall electrode discharge capacity by 35%, to 68 mAh/g_cathode (205 mAh/g_LFP) (Figure 4). Moreover, although the reduction of LZC is characterized by a germination-growth like response, the LFP discharge plateau at 3.4 V exhibited no signs of additional polarization. This demonstrates that although LZC can be advantageously used as a second active material it appears to maintain its electrolyte properties.
• The combined electroactivity of LZC and LFP consistently surpassed that of LFP alone by 22% even after 110 cycles (Figure 5). Furthermore, the 94% capacity retention of LFP confirms the enduring conduction property of this redox electrolyte.
• Notably, increasing the LFP loading from 33 to 50 wt% and areal capacity from 1.4 to 4.1 mAh/cm², yielded competitive gains (+17%) in specific capacity per mass of LZC. In fact, triggering the LZC redox activity allows to reach the same capacity at five times higher C-rate (Figure 6 representing the capacity normalized by: a) total mass of cathode; b) mass of LFP; c) area of the cell in function of the cycle number).
• Interestingly, the initial oxidation process is characterized by an irreversible decrease in cell pressure despite Li alloying occurring at the negative electrode along with a significant increase in cell impedance (Figure 7 showing the variation of a half-cell stack pressure bearing LZC/P2SWNT composite cathode during cycling and Figure 8a showing the EIS of half-cell with LZC/P2SWNT composite cathode at open-circuit voltage (OCV), the first charge being at 5V vs Li⁺/Li and subsequent discharge to 2.5 V vs Li⁺/Li). These results are likely indicative of substantial structural and morphological changes. However, strikingly, electrochemical impedance spectroscopy measurements of the reduced sample (Figure 8b showing the EIS of half-cell with LZC/P2SWNT composite cathode in the discharged state (at 2.5V vs Li+/Li) after 1^st, 2^nd and 10^th cycle) indicates that cell impedance reverts to a value close to the pristine state. This unexpected property is a factor enabling LFP-based electrode to operate without additional polarization, regardless of whether the electrochemical activity of LZC is triggered (Figure 4).
• For the anode-less cell, triggering LZC electroactivity resulted in an astounding 112% capacity increase after 10 cycles, accompanied by significantly improved Coulombic efficiencies (Figure 9). Equally noteworthy is the effect on cell safety and longevity, as dendrite-related irregularities were deferred from the second discharge to the tenth cycle (Figures 10 showing the 1^st (a), 2^nd (b), 5^th (c) and 10^th (d) galvanostatic charge/discharge cycle of the anode-less cell in 2.5V - 4 V and 2.5 - 5 V voltage range).
• These findings indicate the potential of halide electrochemistry for in situ deposition of a thin and uniform layer of lithium metal, offering a solution to common issues encountered in anode-less cell configurations.
• Similarly, a micro-Si based full cell assembled at 4.1 mAh/cm², shows a 37% capacity loss during the initial cycle, coupled with low Coulombic efficiency (Figure 11). However, upon triggering LZC electroactivity, a remarkable 50% enhancement in discharge specific capacity was achieved. These improvements stem from two primary factors: the additional discharge capacity associated with LZC activity, and the heightened activity of LFP (Figure 11, Figure 12 showing the increase in the discharge capacity upon triggering the oxidation of LZC on cycle 4 and resuming to cycling without oxidation of LZC on cycle 5). This underscores the reparative capability of halide electrolytes in Si-based full cells, suggesting that LZC can replenish the loss of active Li/e-.
• Regardless of the Li_xZrCl_4+x (LxZC) composition (x = 0.4, 0.59, 0.8, 1,2 and 3), partially reversible faradic activity was observed at approximately 4V vs. Li⁺/Li (Figure 13a cyclic voltammograms of the different LixZC). This reversible capacity thus enables the enhancement of the positive LFP electrode's capacity, while the irreversible capacity is transferred to the negative electrode, compensating for irreversible electron loss. By leveraging the chloride redox chemistry, the non-stoichiometric composition of the halide electrolyte thus becomes a new and strong parameter for balancing the two electrodes. Indeed, reversible and irreversible capacities show linear and parabolic relationships with the LxZC composition, respectively and the ratio of these two capacities increases by nearly 200% from x=0.4 to x=3 (Figure 13b, showing the variations of reversible and total (reversible + irreversible) capacities referred to as "y" as well as the ratio of y_total/y_reversible against the compositions of the halide Li_xZrCl_4+x solid electrolytes. y_total and y_reversibie correspond to the average value over the 5 first cycles).

Claims (10)

1. - A method for activating an all-solid-state battery comprising a halide electrolyte having the formula:
   
           Li_{[(x-y)+a+b+c+p]}Na_yZr_(1-a-b-c)Er_aY_bSc_cCl_{[(4+x)-m-n-o-p]}I_mBr_nF_oO_p
   
   in which:

   - $$0.2\le \mathrm{x}\le 6;$$

   - $$0\le \mathrm{y}\le 6;$$

   - $$0\le \mathrm{a}\le 1;$$

   - $$0\le \mathrm{b}\le 1;$$

   - $$0\le \mathrm{c}\le 1;$$

   - $$0\le \mathrm{m}\le 1;$$

   - $$0\le \mathrm{n}\le 1;$$

   - $$0\le \mathrm{o}\le 1;$$

   - $$0\le \mathrm{p}\le 2;$$

   - with 0 ≤ a+b+c ≤ 1,

   - with (4+x) > (m+n+o+p),

   characterized in that the battery is activated by charging it at a voltage greater than 4.0 V vs Li⁺/Li.
2. - The method according to claim 1, characterized in that the battery is activated by charging it at a voltage greater than or equal to 4.5 V vs Li⁺/Li, preferably greater than or equal to 5.0 V vs Li⁺/Li.
3. - The method according to one of claims 1 or 2, characterized in that the halide electrolyte is selected from: Li_xZrCl_4+x, wherein x is a number from 0.4 to 6, being in particular Li_0.4ZrCl_4.4, Li_0.59ZrCl_4.59, Li_0.8ZrCl_4.8. LiZrCl₅, Li₂ZrCl₆, Li₃ZrCl₇, Li₄ZrCl₈, Li₅ZrCl₉ and Li₆ZrCl₁₀.
4. - The method according to anyone of claims 1 to 3, characterized by the fact that the battery has a cathode, the active material of which is selected from LiFePO₄, LiMn₂O₄, Li₃V₂(PO₄)₃, LiVOPO₄, Li₂VOPO₄F and Li_a'Ni_x'Mn_y'Co_z'O₂ with x'+y'+z'=1 and 0 < a' < 1.2.
5. - The method according to anyone of claims 1 to 4, characterized by the fact that the battery is anode less.
6. - The method according to anyone of claims 1 to 4, characterized by the fact that the battery has an anode, the active material of which is selected from lithium metal, lithium-indium alloy, silicon, tin or carbon intercalation materials.
7. - The method according to claim 6, characterized by the fact that the lithium-indium alloy is LiIn_x" with 0 < x" ≤ 3.
8. - The method according to anyone of claims 6 and 7, characterized by the fact that the intercalation materials are selected from graphite and hard carbon.
9. - The method according to anyone of claims 1 to 8, characterized by the fact that the halide electrolyte is under the form of a layer.
10. - The method according to claim 9, characterized by the fact that the battery further comprises a second electrolyte layer comprised of sulfide electrolyte, such as -Li₆PS₅Cl.